Homologous recombination is a mechanism that is essential to allow cells to tolerate DNA damage produced by various DNA damaging agents. The long term goals are to define the mechanisms by which agents such as cisplatin, nitric oxide and methylators induce the formation of DNA double-strand breaks and their repair. We have demonstrated, for the first time, using single cell microgel electrophoresis that cisplatin induces the formation of such breaks and it is proposed in the first aim to examine the other agents for their ability to do so and to examine the role of DNA replication in the process. In the second aim, a new class of mutant cells which are dependent on homologous recombination for survival will be sought and characterized especially for their response to DNA damaging agents. Loss of DNA mismatch repair in tumor cells results in drug resistance while, conversely, mismatch repair proficiency leads to drug sensitization. The third aim is based on the finding that the C-terminal end of a key mismatch repair protein, MutS, is needed for drug sensitization and experiments are proposed to determine if the multimeric state of MutS is responsible. We recently showed that cisplatin-induced recombination and double-strand break repair require DNA polymerase I and in the fourth aim, we will determine its role in these processes by inactivating either its exonuclease or polymerase activities. We can use only E. coli for these studies because more is known about DNA replication, repair and recombination than in any other organism and because of the ability to construct multiple mutations in its genome. This basic research impacts several areas of clinical relevance, including mechanisms by which antitumor agents kill cells and how drug-resistant tumors emerge in cancer chemotherapy. It also impacts the mechanism by which pathogenic bacteria become resistant to antibiotics and how this resistance is disseminated. [unreadable] [unreadable] [unreadable]